Salidroside Reduces Cold-Induced Mucin Production by Inhibiting TRPM8 Activation

INTERNATIONAL JOURNAL OF MOLECULAR MEDICINE 32: 637-646, 2013

Salidroside reduces cold-induced mucin production by inhibiting TRPM8 activation

QI LI1, XIANG-DONG ZHOU1, VICTOR P. KOLOSOV2 and JULIY M. PERELMAN2

1Department of Respiratory Medicine, Second Affiliated Hospital, Chongqing Medical University, Chongqing 400010, P.R. China; 2Far Eastern Scientific Center of Physiology and Pathology of Respiration, Siberian Branch, Russian Academy of Medical Sciences, Blagoveschensk 675000, Russia

Received April 18, 2013; Accepted June 28, 2013

DOI: 10.3892/ijmm.2013.1434

Abstract. Salidroside is an effective component of the tradi- gene expression in the absence of CREB activity. The results tional Chinese herb, Rhodiola rosea, that is known to have revealed that the cells treated with either CREB siRNA or sali- the ability to protect individuals from cold attacks. In the droside expressed low levels of TRPM8 mRNA and protein. present study, we investigated the effects of salidroside on These results indicate that salidroside reduces MUC overpro- respiratory epithelial cells exposed to cold temperatures. We duction induced by cold stimuli and that salidroside exerts its wished to determine whether salidroside exerts any effect protective effects by inhibiting TRPM8 activation, mainly by on cold-induced mucin (MUC) production and the possible decreasing CREB activity. mechanisms involved in this process. We incubated HBE16 cells with salidroside, exposed them to a cold stimulus (18˚C), Introduction and assayed the following endpoints: MUC production (the expression of MUC5AC), concentration intracellular of free Cold stimuli are known to exacerbate chronic obstructive calcium ([Ca2+]i), the activation of the transient receptor poten- pulmonary disease (COPD) and have been shown to promote tial melastatin 8 (TRPM8) channel and the cAMP response mucin (MUC) hypersecretion (1). Excessive MUC produc- element-binding protein (CREB). Our results revealed a tion leads to airway obstruction and enhances inflammation. significant increase in the [Ca2+]i concentration, as well as in Previous studies have mainly focused on secondary inflamma- TRPM8 and CREB expression in the cold-stimulated cells. tory processes (2) and bacterial (3) or viral infections (4,5) that MUC5AC expression was also increased. Treatment of the cells occur in the airways following exposure to cold temperatures; with salidroside at concentrations of 50 and 100 µM decreased however, the molecular mechanisms behind cold‑induced the [Ca2+]i concentration, with a maximal effect detected in mucus secretion have not yet been fully elucidated. Human the cells treated with 100 µM salidroside. The expression of organs express ion channels of the transient receptor poten- TRPM8 and TRPM8 channel conductivity were also repressed tial (TRP) family, including some ion channels that respond by salidroside; salidroside decreased the high levels of CREB at distinct temperature thresholds (6‑9). TRP melastatin 8 activity and phosphorylation observed in the cold‑stimulated (TRPM8) is a member of this family that is activated by cells. Furthermore, we transfected the cold‑stimulated cells temperatures ranging from 8 to 25˚C. TRPM8 is a ligand‑gated with CREB small interfering RNA (siRNA) to analyze TRPM8 cation channel with moderate to high selectivity for calcium ions (10,11). TRPM8 can be activated by cold temperatures and by cooling agents, such as menthol, eucalyptol and icilin. In Correspondence to: Professor Xiang-Dong Zhou, Department a previous study, we found that TRPM8 plays a critical of Respiratory Medicine, Second Affiliated Hospital, Chongqing role in cold‑induced MUC secretion in the human respira- Medical University, No. 74 Linjiang Road, Chongqing 400010, tory system (12). Human airway epithelial cells exposed to P. R. Ch i na cold temperatures exhibited activated TRPM8 channels, E-mail: [email protected] and MUC5AC expression was upregulated through the TRPM8‑Ca2+‑phospholipase C (PLC)‑phosphatidylinositol Abbreviations: TRPM8, transient receptor potential melastatin 8; CRE, cAMP response element; ELISA, enzyme-linked immuno bisphosphate (PIP2) pathway. Thus, we hypothesized that sorbent assay; HRP, horseradish peroxidase; MUC, mucin; siRNA, TRPM8 plays a role in cold-induced MUC secretion. In this small interfering RNA study, we aimed to discover methods of inhibiting cold‑induced MUC hypersecretion. Key words: salidroside, mucin, airway, cold stimuli, transient Salidroside (rhodioside) is a glycoside compound receptor potential melastatin 8 primarily found in the plant, Rhodiola rosea, which grows in cold, high‑latitude areas. It is well known that salidroside has pharmacological effects on many pathological conditions, including radiation poisoning (13), inflammation (14), fatigue, 638 LI et al: SALIDROSIDE INHIBITS TRPM8 ACTIVATION aging and hypoxia (15,16). Other studies have found that siRNA transfection. Cells were plated at a density of salidroside exerts its protective effects mainly by repressing 1‑2x106 cells/ml and incubated at 37˚C until the cells were the inflow of intracellular free calcium (Ca2+) ([Ca2+]i) into 60‑80% confluent. CREB siRNA (0.5 µg) were diluted in cells. Considering what is already known about salidroside 100 µl of siRNA transfection medium (solution A); 6 µl and TRPM8, we hypothesized that salidroside has the ability siRNA transfection reagent was diluted in 100 µl of siRNA to protect cells from cold nociceptive stimuli and that it may transfection medium (solution B). The solutions (A + B) were play a positive role in the airway system. In the present study, subsequently mixed gently and incubated at room temperature we explored the regulation of TRPM8 gene transcription in for 45 min. The cells were washed once with 2 ml of siRNA cold-assaulted cells and examined the effects of salidroside on transfection medium followed by the addition of 0.8 ml of the TRPM8 channels and MUC production. We used the HBE16 siRNA transfection reagent mixture (solutions A + B). The human airway epithelial cell line for in vitro studies. The cells were then incubated for 7 h at 37˚C in a CO2 incubator cells were pre-treated with salidroside and exposed to cold and were recovered in 2 ml of DMEM/Ham's F12 medium temperatures; the protein and mRNA levels of MUC5AC were containing 10% FBS for an additional 24 h. The medium was then determined. The expression levels of TRPM8 and the then aspirated and replaced with fresh growth medium for the transcription factor, cAMP response element‑binding protein following assays (see below for details). (CREB), were assayed to investigate whether there were any differences in the levels of these factors following treatment RNA isolation and real-time PCR for MUC5AC and TRPM8 with salidroside. Furthermore, we transfected the cells with mRNA. After the HBE16 cells were harvested, total RNA was CREB small interfering RNA (siRNA) to determine whether extracted from the cells using the Bioteke high-purity total the transcription factor, CREB, is involved in TRPM8 gene RNA extraction kit, and RNA integrity was verified using transcription. 1.5% agarose gel electrophoresis. The absorbance at 260/280 nm

(A260/280) was in the range of 1.8‑2.0. Total RNA was primed Materials and methods with an oligo(dT) primer and reverse transcribed using the PrimeScript RT Reagent kit. Real-time PCR was performed Reagents. Dulbecco's modified Eagle's medium (DMEM)/ using the SYBR Premix EX Taq™ II real‑time PCR kit. The Ham's F12 medium, HEPES, fetal bovine serum (FBS), sequences of the primers used for real-time PCR were as anti‑β‑actin monoclonal antibody, salidroside (>98% purity; follows: MUC5AC (U06711) forward, 5'-CAGCCACGTCCCC National Institute for Food and Drug Control), CREB siRNA and TTCAATA-3' and reverse, 5'-ACCGCATTTGGGCATCC-3'; negative control siRNA were purchased from Sigma‑Aldrich TRPM8 (NM024080) forward, 5'-ACTCAGAAGGCTGAGG (St. Louis, MO, USA). Lipofectamine® 2000 and OPTI‑MEM TACA-3' and reverse, 5'-TTCAGTCGGAGTCTCACTCT-3'; Reduced Serum Medium were purchased from Invitrogen and GAPDH (BC026907) forward, 5'-GAAGGTGAAGGT (San Diego, CA, USA). The rabbit anti‑TRPM8 polyclonal CGGAGT-3' and reverse, 5'-GAAGATGGTGATGGGA antibody was purchased from Abcam (Cambridge, MA, USA) TTTC-3'. GAPDH was used as the loading control. Relative and the horseradish peroxidase (HRP)-goat anti-rabbit IgG mRNA expression was determined by comparing it to a antibody was purchased from Santa Cruz Biotechnology, standard curve. Inc. (Santa Cruz, CA, USA). CREB-Luc reporter constructs were purchased from Biocat (Catalonia, Spain). The Dual Enzyme-linked immunosorbent assay (ELISA) for MUC5AC Luciferase Reporter Gene Assay kit was purchased from protein. MUC5AC protein levels in the cell lysates and BioTime Technology (Beijing, China). The pRL-TK Renilla supernatants were measured using ELISA. Cell lysates at Luciferase Reporter Vector was purchased from Promega multiple dilutions were prepared with phosphate-buffered (Madison, WI, USA). The high‑purity total RNA extraction kit saline (PBS). Cell culture supernatants were collected for was purchased from Bioteke Biotechnology (Beijing, China). this assay. In addition, 50 µl of each sample were incubated The PrimeScript RT Reagent kit was purchased from Takara with bicarbonate‑carbonate buffer (50 µl) at 40˚C in a 96‑well (Dalian, China). plate until dry. The wells were washed 3 times with PBS and blocked with 2% bovine serum albumin (BSA) for 1 h at room Cell culture. HBE16 human airway epithelial cells were temperature. The wells were then incubated with mouse anti- plated in 6‑well plates at a concentration of 5‑6x105 cells/well human MUC5AC monoclonal antibody (45M1) (10 µg/ml; in 2 ml DMEM/Ham's F12 medium that contained 10% FBS. Neomarkers Inc., Fremont, CA, USA) diluted in PBS containing

The cells were then incubated at 37˚C and 5% CO2. The 0.05% Tween‑20 for 1 h at room temperature. After washing culture medium was changed to growth factor‑free medium 3 times, the wells were incubated with HRP‑conjugated goat prior to the start of each experiment. Cell supernatants anti‑mouse IgG (1 µg/ml) for 1 h. HRP was developed with and lysates were collected and assays were performed, as 3,3',5,5'-tetramethylbenzidine (TMB) peroxidase solution, described below. quenched with 1 M H2SO4 and the absorbance at 450 nm was measured. Cell viability measurements. Confluent HBE16 cells were incubated with 10, 50, 100, or 200 µM salidroside for 24 h. We Western blot analysis for TRPM8 and phosphorylated CREB used the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium (p-CREB) protein levels. Total protein was determined by bromide (MTT) reduction method to detect cell viability and BCA analysis. The cells were lysed in lysate buffer, disrupted to select the appropriate dose of salidroside for the subsequent on ice for 20 min, and then centrifuged at 12,000 rpm for experiments. 15 min at 4˚C to remove the nuclei and unbroken cells. The INTERNATIONAL JOURNAL OF MOLECULAR MEDICINE 32: 637-646, 2013 639 total protein concentration in the cell lysates was determined The 340/380 nm fluorescence intensity ratio of each sample by BCA assay. Equal amounts of protein were diluted in SDS is abbreviated as R. The 380 nm fluorescence intensity of sample buffer and boiled for 5 min. Proteins were resolved Fura‑2 under CA2+-free and Ca2+ saturation conditions are on 6% SDS gels. After running, the gels were equilibrated abbreviated as Fmin and Fmax, respectively. As the controls, in transfer buffer containing 25 mM Tris‑HCl, 192 mM 40 µl 2.5% Triton X‑100 were added to the cells to detect glycine and 20% methanol (pH 8.3). The proteins were then the maximum fluorescence intensity ratio (Rmax) and 40 µl transferred by electrophoresis onto polyvinylidene difluoride 250 mM Ca2+ chelator EGTA were added to the cells to detect (PVDF) membranes. The membranes were incubated in the minimum fluorescence intensity ratio (Rmin). The [Ca2+] 5% milk dissolved in PBS and 0.05% Tween‑20 for 1 h at i concentrations were calculated using the formula [Ca2+] room temperature and subsequently incubated with rabbit i = Kd[(R - Rmin)/(Rmax - R)](Fmin/Fmax). anti‑TRPM8 (diluted 1:200) or rabbit anti‑p‑CREB (Ser‑129) polyclonal antibodies (diluted 1:500) overnight at 4˚C. After CREB activity assay. CREB activity was determined using washing 3 times, the membranes were incubated with an the CREB-Luciferase Reporter Assay System (Biocat). HRP-labeled goat anti‑rabbit IgG antibody (diluted 1:1,000) for The CREB‑Luc and pRL‑TK Renilla Luciferase Reporter 2 h at room temperature. Enhanced chemiluminescence (ECL) plasmids were transfected into the HBE16 cells using and autoradiography were performed to visualize TRPM8 and Lipofectamine® 2000. Eighteen hours after transfection, the p‑CREB protein. The membranes were also blotted for β‑actin cells were pre‑incubated with 100 µM salidroside for 30 min and α‑tubulin to ensure equal protein loading. and exposed to cold temperatures (18˚C). CREB activity was assayed in accordance with the manufacturer's instructions Immunofluorescence for TRPM8. Cells were fixed with 4% provided with the Luciferase Reporter Gene Assay kit. The paraformaldehyde and subsequently permeabilized with results were expressed as detected luciferase relative light 0.1% Triton X‑100 in PBS for 30 min. The cells were rinsed, units (RLU)/Renilla luciferase RLU. blocked in 1% BSA plus 1% normal goat serum and incubated with rabbit anti‑TRPM8 (diluted 1:500) overnight at 4˚C. Statistical analysis. Data analysis was performed using the After 3 washes of 10 min each in 0.1% Triton X‑100/PBS, the SPSS 17.0 statistical software package (SPSS, Inc., Chicago, cells were incubated with FITC‑conjugated fluorescent goat IL, USA). All in vitro experiments using cell lines included anti‑rabbit IgG (1:500) for 2 h at room temperature followed samples from at least 6 separate wells and 2‑3 independent by cellular nuclear staining with 4'6-diamidino-2-phenyl- experiments. Comparisons of >2 groups were made using the indole (DAPI). The samples were examined using a Leica Student's t‑test or a one‑way analysis of variance (ANOVA) inverted TCS‑SP2 confocal microscope that was equipped followed by Bonferroni's analysis. A P-value <0.05 was with appropriate fluorescence filters. considered to indicate a statistically significant difference.

Determination of TRPM8 channel conductivity. Whole‑cell Results patch‑clamp experiments were performed using an EPC‑10 amplifier and IGOR Pro 4.0 software for data acquisition and Viability of HBE16 cells following exposure to salidroside. The analysis. Frequencies of 10 KHz were selected for sampling cells were incubated with 10, 50, 100, or 200 µM salidroside and 2 KHz for filtration. The extracellular solution used for for various periods of time to determine cellular prolifera- electrophysiological recordings consisted of 140 mM NaCl, tion rates following exposure to salidroside. Salidroside at a

5 mM KCl, 2 mM CaCl2, 1 mM MgCl2, 0.3 mM Na2HPO4, concentration of 200 µM inhibited cell proliferation after 16 h. 0.4 mM KH2PO4, 4 mM NaHCO3, 5 mM glucose and 10 mM Following incubation with 10, 50, or 100 µM salidroside for HEPES (the pH was adjusted to 7.4 using NaOH). The intra- 24 h, the viability of these cells did not differ significantly cellular solution filled into patch-clamp pipettes contained from that of the non‑treated cells (Fig. 1). These results indi-

140 mM KCl, 1 mM MgCl2, 2.5 mM CaCl2, 4 mM EGTA and cate that lower doses of salidroside do not affect cell viability. 10 mM HEPES. The calculated free Ca2+ concentration in this Therefore, we selected to use the concentrations of 50 and solution was 150 nM, and the pH was adjusted to 7.2 using 100 µM for the following experiments. KOH as previously described (17). The borosilicate glass patch pipettes used for the whole‑cell recordings had resistances of Salidroside decreases cold‑induced MUC5AC expression. 2‑4 MΩ when filled with intracellular solutions. To investigate the effects of cold temperatures on MUC5AC expression in airway epithelial cells, we exposed the HBE16 Ca2+ influx detection.The cells were exposed to cold tempera- cells to cold temperatures (18˚C) for various periods of time. tures (18˚C) or pre-treated with salidroside and collected at MUC5AC mRNA levels were upregulated after 1 h of expo- different time intervals. We resuspended the cells in D-Hanks sure to cold temperatures, with the concentration of MUC5AC buffer containing 0.2% FBS (106 cells/ml), removed 1 ml of protein in the cell lysates and culture supernatants increasing the cell suspension to an Eppendorf tube and then added 2 µl as well. The mRNA and protein levels of MUC5AC continued Fura‑2/AM stock solution to the cells followed by incubation to increase, reaching a plateau at 8 h. These results indicate with shaking at 37˚C for 60 min. The cells were centrifuged at that exposure to cold temperatures actively induces MUC5AC 1,300 rpm for 6 min, washed twice with pre-heated D‑Hanks expression. The cells were then incubated with 50 µM sali- buffer containing 0.2% BSA and resuspended in D‑Hanks droside prior to exposure cold temperatures. The MUC5AC buffer. The fluorescence intensity was then observed using protein and mRNA levels in the salidroside-treated cells were a dual wavelength spectrophotometer (Hitachi F‑3000). lower than those observed in the cold‑stimulated group at 6 h. 640 LI et al: SALIDROSIDE INHIBITS TRPM8 ACTIVATION

Figure 1. The viability of the HBE16 cells was not affected by 100 µM salidroside. HBE16 cells were incubated with various concentrations of salidroside and the growth rates of the cells in each group were measured by MTT assay at 4, 6, 8, 12, 16 and 24 h of incubation. Sal, salidroside.

Treatment of the cells with 100 µM salidroside for 8 h significantly attenuated the cold‑induced intracellular synthesis and secretion of the MUC5AC protein (Fig. 2), indicating that salidroside time- and dose‑dependently reduced MUC5AC expression induced by cold stimulation.

Salidroside decreases ITRPM8 membrane currents. In a previous study, we found that cold temperatures induce TRPM8 activation (12). We further assayed the activation of TRPM8 in cold‑stimulated HBE16 cells by patch-clamping the cells at various time points following exposrue to cold temperatures. The current density of cold‑stimulated TRPM8 channels increased at 15 sec and reached a plateau at 6 min. In the salidroside‑treated cells, TRPM8 activation began to decrease after 40 sec. Salidroside significantly decreased the current through cold‑stimulated TRPM8 channels at 10 min of recording in comparison with the untreated cold‑stimulated channels (P<0.05). The 100 µM concentration of salidroside was more effective at reducing currents than the 50 µM concentration, showing maximal effects at 20 min (Fig. 3). These results indicate that TRPM8 activation is involved in Figure 2. Salidroside downregulates cold‑induced MUC5AC expression. cold‑stimulated MUC expression and that salidroside inhibits (a) Cells were treated with various concentrations of salidroside (50 or currents through TRPM8, suggesting a previously unknown 100 µM) prior to exposure to cold temperatures. The MUC5AC mRNA levels in each group were detected by real‑time PCR at various time points (1, 4, 6, property of salidroside. 8, 10, 12 h) as described in Materials and methods. The data shown are the means ± SD from 6 different samples. (b) MUC5AC protein levels in cell Salidroside reduces cold‑induced TRPM8 expression in lysates. Cells were treated with salidroside or exposed to cold temperatures HBE16 cells. To further confirm the inhibitory effects of sali- (18˚C) as described in (a) and subsequently harvested, lysed and analyzed by ELISA. (c) MUC5AC protein levels in cell supernatants. Cells were droside on TRPM8, we investigated the mRNA and protein treated as described in (a) and cell supernatants were collected and tested expression of TRPM8 following incubation with salidroside. for MUC5AC protein levels by ELISA. *P<0.05 compared with the control △ ▲ The cells were pre-treated with salidroside and then exposed group; P<0.01 compared with the control group; P<0.05 compared with # to cold temperatures (18˚C). The TRPM8 mRNA and protein the 50 µM salidroside-treated group; P<0.01 compared with the 100 µM salidroside-treated group. P‑values were calculated using the paired Student's levels were then detected by real‑time PCR, western blot t‑test. Sal, salidroside. analysis and immunofluorescence assay, respectively. The cells were exposed to cold temperatures (18˚C) for 6 h; exposure to cold temperatures significantly increased TRPM8 protein and mRNA levels compared with the normal control group in a time‑dependent manner. The cells pre-treated with (P<0.05), and the cellular TRPM8 levels continued to increase salidroside prior to exposure to cold temperatures had lower INTERNATIONAL JOURNAL OF MOLECULAR MEDICINE 32: 637-646, 2013 641

Figure 3. Salidroside inhibits cold‑induced membrane currents through transient receptor potential melastatin 8 (TRPM8) channels in HBE16 cells. Cells were pre‑incubated with salidroside (50 and 100 µM) and then exposed to cold temperatures (18˚C). The TRPM8 currents at various time points following exposure to cold temperatures were determined by a patch clamp. *P<0.05 compared with the control group; △P<0.05 compared with the 100 µM salidroside-treated group; ▲P<0.01 compared with the 100 µM salidroside-treated group. P‑values were calculated using the paired Student's t‑test. Sal, salidroside.

TRPM8 expression levels than the non‑treated cells (P<0.05). The high dose of salidroside (100 µM) showed more potent inhibitory effects on increased TRPM8 expression than the low dose (50 µM) (Figs. 4 and 5). These results indicate that salidroside represses TRPM8 not only by reducing TRPM8 activation but also by inhibiting TRPM8 mRNA and protein expression.

Salidroside inhibits [Ca2+]i in HBE16 cells. To investigate the mechanisms underlying the salidroside-mediated TRPM8 inhibition, we assayed the free calcium influx in the cells, which is crucial to TRPM8 activation and is known to play Figure 4. Salidroside decreases transient receptor potential melastatin 8 a role in many other salidroside-mediated protective events. (TRPM8) expression in cold‑stimulated cells. HBE16 cells were treated with The HBE16 cells were loaded with Fura‑2/AM to examine the 2 different concentrations of salidroside (50 and 100 µM) for 6 h. (a) The level of TRPM8 mRNA in the HBE16 cells following treatment with salidro- intracellular calcium concentration. When the HBE16 cells side was detected by real‑time PCR. (b) The level of TRPM8 protein in the were exposed to cold temperatures (18˚C), we observed that HBE16 cells following treatment with salidroside was determined by western the [Ca2+]i concentration began to increase after 4 min and blot analysis. TRPM8 expression levels were determined using densitometry continued to increase, reaching a plateau at 10 min, compared analysis. Protein levels were calculated as the ratio of TRPM8 to β‑actin and normalized to the control. Data shown are the means ± SD from 6 different with the normal group (P<0.01). In the HBE16 cells incubated samples. *P<0.05 compared with the control group; △P<0.01 compared with with salidroside prior to exposure to cold temperatures, the the cold-stimulated (18˚C) group; ▲P<0.05 compared with the cold-stimulated increase in the [Ca2+]i concentration was significantly attenu- (18˚C) group. P‑values were calculated using the paired Student's t‑test. Sal, ated at 8 min. Furthermore, pre-treatment of the cells with salidroside. both concentrations of salidroside (50 and 100 µM) decreased the [Ca2+]i concentration in a dose‑dependent manner (Fig. 6).

Salidroside decreases TRPM8 expression by inhibiting CREB expression. Following the incubation of the HBE16 cells activity in HBE16 cells. Previous studies have reported that under cold conditions for 6 h, the mRNA and protein levels the TRPM8 promoter may have a CREB binding site (18). To of TRPM8 increased when compared with the control cells determine whether CREB is involved in TRPM8 activation, kept at warm temperatures. When the cells were transfected we first assayed the expression and activity of CREB in the with CREB siRNA prior to exposure to cold temperatures cells under cold assault, and then transfected the cells with (18˚C), the TRPM8 mRNA and protein levels were consider- CREB siRNA to investigate the effects of CREB on TRPM8 ably lower than the levels observed in the non‑transfected cells 642 LI et al: SALIDROSIDE INHIBITS TRPM8 ACTIVATION

Figure 5. Salidroside and cAMP response element-binding protein (CREB) small interfering RNA (siRNA) decrease the expression of transient receptor potential melastatin 8 (TRPM8) protein in HBE16 cells. HBE16 cells were treated with 2 different concentrations of salidroside (50 and 100 µM) or transfetced with CREB siRNA. The protein expression was assayed by immunofluorescence. (a) Untreated HBE16 cells; (b) cold-stimulated (18˚C) group; (c) cold- stimulated (18˚C) + 50 µM salidroside-treated group; (d) cold-stimulated (18˚C) + 100 µM salidroside-treated group; (e) cold-stimulated (18˚C) + CREB siRNA‑transfected group; (f) cold-stimulated (18˚C) + control siRNA‑transfected group. The samples were run in triplicate and in 3 separate experiments. Data are expressed as the means ± SE, n=6.

(Figs. 5 and 7). Exposure to cold temperatures also increased transcription and that salidroside represses TRPM8 expression the levels of p‑CREB. The cells transfected with CREB siRNA by inhibiting CREB activity. showed lower levels of p‑CREB, an effect similar to the one observed in the salidroside-treated cells (P<0.05 compared Discussion with the cold-stimulated (18˚C) cells (Fig. 8A). We addition- ally transfected the cells with a CREB-Luc Reporter plasmid Mucus overproduction is a clinical hallmark of acute exacer- to measure the activity of CREB in the cold‑stimulated cells. bation of COPD. One of the important aggravating factors is The cells exposed to cold temperatures (18˚C) showed higher exposure to cold temperatures, which can cause mucociliary RLU levels than the normal control cells, while the cells incu- dysfunction and the accumulation of inflammatory factors, bated with salidroside prior to exposure to cold temperatures which may ultimately result in a bacterial infection in the exhibited lower RLU levels than the cold‑stimulated cells respiratory system. The effective management of COPD exac- not treated with salidroside (P<0.01) (Fig. 8B). These results erbation requires a thorough understanding of the underlying suggest that CREB plays an important role in TRPM8 gene pathophysiological mechanisms that shape its clinical manifes- INTERNATIONAL JOURNAL OF MOLECULAR MEDICINE 32: 637-646, 2013 643

Figure 6. Salidroside attenuates cold‑induced increase in intracellular free Ca2+ ([Ca2+]i) concentration in HBE16 cells. Cells were exposed to cold temperatures (18˚C) for 4 min, at which point the [Ca2+]i concentration began to increase. This effect became more significant at 6 min. **P<0.01 compared with the control group. Salidroside attenuated the increase in the [Ca2+]i concentration at 10 min. *P<0.05 compared with the 50 µM salidroside-treated group; #P<0.01 compared with the 100 µM salidroside-treated group. Paired Student's t‑tests were used to calculate the P‑values.

tation. In a previous study, we reported that cold temperatures induce MUC hypersecretion in human airway epithelial cells through a TRPM8‑mediated mechanism (12). TRPM8 is a major cold receptor in the body with variants that are expressed by the human respiratory system (19). TRPM8 is activated by temperatures between 8 and 25˚C (20) and is a member of the TRP family of Ca2+‑permeable, non‑selective cation channels and the TRPM subfamily of TRP proteins (21‑23). We found that TRPM8 is highly expressed in human airway epithelial cells following exposure to cold stimuli. Functional assays revealed that TRPM8 activation significantly reduced MUC secretion induced by cold temperatures; however, the interven- tion mechanims for TRPM8-related events have not yet been fully elucidated. Rhodiola rosea has long been used in traditional Chinese medicine and it primarily grows at high altitude in thin air and cold weather conditions. Salidroside is the major active ingre- Figure 7. Transfection of cAMP response element-binding protein (CREB) dient isolated from the plant, Rhodiola rosea, and possesses small interfering RNA (siRNA) represses cold‑induced transient receptor the ability to enhance the resistance of the body to fatigue and potential melastatin 8 (TRPM8) expression. HBE16 cells were transfected cold (24). Salidroside has also been reported to have potent with CREB siRNA prior to exposure to cold temperatures (18˚C|). Cells were then cultured for an additional 6 h, at which time TRPM8 mRNA levels were antioxidant and anti‑aging properties (25). Salidroside has been assayed by real‑time PCR and protein levels were measured by western blot shown to reduce H2O2‑induced cell apoptosis in SH‑SY5Y analysis. (a) TRPM8 mRNA expression following transfection with CREB human neuroblastoma cells (26,27). However, the potential siRNA. (b) TRPM8 protein expression following transfection with CREB * ▲ effects of salidroside on cold‑induced airway inflammation siRNA. P<0.05 compared with the control group. P<0.01 compared with the cold-stimulated (18˚C) group. P‑values were calculated using the paired have not yet been fully elucidated. The respiratory epithelium Student's t‑test. Sal, salidroside. communicates directly with the external environment. When cold air is inhaled, the respiratory tract loses a considerable amount of heat. This heat loss may be followed by a series of stress reactions. Cold air‑induced respiratory symptoms include perturbed mucociliary function, placing them at a higher risk MUC oversecretion and inflammation, and these phenomena of developing nasal obstruction and cough (29). Considering are common in countries or areas with a cold climate (28). It has the known properties of salidroside, we hypothesized that it also been reported that people living at high latitudes exhibit may play a role in regulating homeostasis in the airway system. 644 LI et al: SALIDROSIDE INHIBITS TRPM8 ACTIVATION

Figure 8. Salidroside reduces cold‑induced cAMP response element-binding protein (CREB) phosphorylation and activity. (a) HBE16 cells were cultured in the presence of 100 µM salidroside or transfected with CREB small interfering RNA (siRNA) and then exposed to cold temperatures (18˚C). p‑CREB and α‑tubulin protein levels were measures by western blot analysis. To quantitatively determine p‑CREB levels, we performed densitometry analysis for each sample. Protein levels are calculated as the ratio of p‑CREB to α‑tubulin and normalized to the control. Data shown are the means ± SD from 6 different samples. (b) Cells were transfected with a CREB‑Luc Reporter plasmid to measure CREB activity. *P<0.05 compared with the control group; △P<0.01 compared with the cold- stimulated (18˚C) group; ▲P<0.05 compared with the cold-stimulated (18˚C) group. P‑values were calculated using the paired Student's t‑test. Sal, salidroside.

In this study, we found that HBE16 cells exposed to cold of Ca2+ leads to desensitization of TRPM8, a process that is temperatures exhibited a marked increase in MUC5AC levels mediated through the Ca2+-dependent activation of protein and TRPM8 activation. By contrast, cells pre-treated with kinase C (PKC) (34,35). TRPM8 is also regulated by a number various concentrations of salidroside (10, 50 and 100 µM) of kinases and second messengers, such as PLC, PIP2 (36,37), prior to the cold assault exhibited a dose- and time‑dependent phospholipase A2 (PLA2), lysophospholipids (LPLs) (38,39), decrease in MUC5AC expression, indicating that salidroside arachidonic acid (AA) (40) and cAMP (41). is capable of protecting HBE16 cells from cold-induced MUC CREB is a 43‑kDa leucine zipper transcription factor that oversecretion. Furthermore, we found that the [Ca2+]i concen- belongs to the activating transcription factor (ATF) subfamily tration rapidly and robustly increased in the cold‑stimulated of the basic-region leucine zipper (bZIP) family. CREB cells, reaching a plateau at 6 to 10 min following exposure primarily responds to the cAMP signaling pathway, binding to cold temperatures. The cells pre-treated with salidroside to CREs to increase or decrease the transcription of certain showed decreased levels of [Ca2+]i and TRPM8 activation, genes (42). CREB‑mediated transcription regulates various with the [Ca2+]i concentration decreasing as early as 1 min cellular responses, including intermediary metabolism, following incubation with salidroside and reaching a minimal neuronal signaling, cell proliferation, apoptosis and other level at 2 h post-treatment. High doses of salidroside inhibited molecular events. The CREB protein contains an N‑terminal the inflow of Ca2+ almost completely. transaction domain (kinase-induced domain), a basic region [Ca2+]i is an important second messenger in various types and a leucine zipper motif at the C‑terminus. The transaction of cells, regulating functional and physiological cellular activi- domain includes a phosphorylation site at Ser133 that is recog- ties, such as contraction, secretion, signal transmission and nized by certain protein kinases (43). CREB can be activated transcriptional regulation. Under normal circumstances, the by a variety of stimuli, including peptide hormones, growth extracellular concentration of calcium is usually 0.1‑10 mM factors, forskolin, phorbol esters and a number of protein and the intracellular calcium concentration is 10-7‑10 -8 M, kinases, including protein kinase A (PKA), pp90 ribosomal much lower than the calcium concentrations in the extracel- S6 kinase (pp90RSK), and Ca2+/calmodulin‑dependent protein lular environment. Ca2+‑dependent cellular signaling pathways kinases (CaMKs) (44,45). are activated by alterations in the levels of [Ca2+]i. When cells It has been reported that the TRPM8 core promoter are exposed to stimuli such as inflammation (30), hypoxia (31) region contains a CRE, which is possibly located between and cold temperatures (32), there is an influx of extracellular base pairs -1,417 and -1,406. It may play an important role calcium or a release of calcium from the cellular endoplasmic in promoting the transcription of TRPM8 (18,46). Therefore, reticulum, either of which overloads the [Ca2+]i level. The we further examine whether the transcription factor, CREB, [Ca2+]i concentration plays an important role in the activation is involved in TRPM8 expression. We found that the expres- of TRPM8. Studies have also shown that TRPM8 mediates sion of p-CREB protein and CREB activity were significantly transmembrane Ca2+ flux (33). However, the excessive inflow elevated in the cells exposed to cold temperatures compared INTERNATIONAL JOURNAL OF MOLECULAR MEDICINE 32: 637-646, 2013 645 with the control cells. CREB phosphorylation and activity 12. Li MC, Li Q, Yang G, Kolosov VP, Perelman JM and Zhou XD: Cold temperature induces mucin hypersecretion from normal levels were lower in the cells pre-treated with salidroside than human bronchial epithelial cells in vitro through a transient the untreated cells. While TRPM8 expression decreased in the receptor potential melastatin 8 (TRPM8)‑mediated mechanism. cells transfected with CREB siRNA, the expression of TRPM8 J Allergy Clin Immunol 128: 626‑634, 2011. 13. Li YR, Cao W, Guo J, Miao S, Ding GR, Li KC, Wang J and in the cells transfected with CREB siRNA and incubated with Guo GZ: Comparative investigations on the protective effects of salidroside prior to exposure to cold temperatures decreased rhodioside, ciwujianoside‑B and astragaloside IV on radiation even more significantly. These results indicate not only that injuries of the hematopoietic system in mice. Phytother Res 25: 644‑653, 2011. CREB activation directly contributes to TRPM8 expression, but 14. Díaz Lanza AM, Abad Martínez MJ, Fernández Matellano L, also that salidroside inhibits TRPM8 expression by decreasing Recuero Carretero C, Villaescusa Castillo L, Silván Sen AM and the activation and activity of CREB. Bermejo Benito P: Lignan and phenylpropanoid glycosides from Phillyrea latifolia and their in vitro anti inflammatory activity. The known chemical antagonists of TRPM8 are capsaz- Planta Med 67: 219‑223, 2001. epine, N-(4-tert-butylphenyl)-4-(3-chloropyridin-2-yl) 15. Kelly GS: Rhodiola rosea: a possible plant adaptogen. Altern tetrahydropyrazine-1(2H)-carboxamite (BCTC) and thiol Med Rev 6: 293‑302, 2001. 16. Kucinskaite A, Briedis V and Savickas A: Experimental analysis BCTC. While these are all synthesized chemicals, salidroside of therapeutic properties of Rhodiola rosea L. and its possible is a component of the Chinese herb, Rhodiola rosea, which application in medicine. Medicina (Kaunas) 40: 614‑619, 2004 may be a natural candidate for the treatment of diseases (In Lithuanian). 17. Bavencoffe A, Kondratskyi A, Gkika D, Mauroy B, Shuba Y, involving TRPM8 overexpression. The regulation of TRPM8 Prevarskaya N and Skryma R: Complex regulation of the expression by salidroside is important for the identification of TRPM8 cold receptor channel: role of arachidonic acid release alternative treatment modalities for airway mucus hypersecre- following M3 muscarinic receptor stimulation. J Biol Chem 286: 9849‑9855, 2011. tory diseases induced by cold stimuli. 18. Bidaux G, Roudbaraki M, Merle C, Crépin A, Delcourt P, Slomianny C, Thebault S, Bonnal JL, Benahmed M, Cabon F, Acknowledgements Mauroy B and Prevarskaya N: Evidence for specific TRPM8 expression in human prostate secretory epithelial cells: functional androgen receptor requirement. Endocr Relat Cancer 12: The present study was supported by the National Nature 367‑382, 2005. Science Foundation of China (grant no. 81100003 and 19. Sabnis AS, Shadid M, Yost GS and Reilly CA: Human lung epithelial cells express a functional cold‑sensing TRPM8 variant. 31171346), and the China-Russia Cooperation Research Am J Respir Cell Mol Biol 39: 466‑474, 2008. Foundation (grant no. 31211120168), and the New Teacher 20. McCoy DD, Knowlton WM and McKemy DD: Scraping through Fund for Doctor Station, the Ministry of Education, China the ice: uncovering the role of TRPM8 in cold transduction. Am J Physiol Regul Integr Comp Physiol 300: R1278‑R1287, 2011. (no. 20115503120006), and Chongqing Nature Science 21. Zhang L and Barritt GJ: TRPM8 in prostate cancer cells: a Foundation (grant no. cstc2011jjA10046). potential diagnostic and prognostic marker with a secretory function? Endocr Relat Cancer 13: 27‑38, 2006. 22. Behrendt HJ, Germann T, Gillen C, Hatt H and Jostock R: References Characterization of the mouse cold‑menthol receptor TRPM8 and vanilloid receptor type‑1 VR1 using a fluorometric 1. Giesbrecht GG: The respiratory system in a cold environment. imaging plate reader (FLIPR) assay. Br J Pharmacol 141: Aviat Space Environ Med 66: 890‑902, 1995. 737‑745, 2004. 2. Voynow JA and Rubin BK: Mucins, mucus, and sputum. 23. Andersson DA, Chase HW and Bevan S: TRPM8 activation by Chest 135: 505‑512, 2009. menthol, icilin, and cold is differentially modulated by intra- 3. Zhen G, Park SW, Nguyenvu LT, Rodriguez MW, Barbeau R, cellular pH. J Neurosci 24: 5364‑5369, 2004. Paquet AC and Erle DJ: IL‑13 and epidermal growth factor 24. Tolonen A, Pakonen M, Hohtola A and Jalonen J: Phenylpropanoid receptor have critical but distinct roles in epithelial cell mucin glycosides from Rhodiola rosea. Chem Pharm Bull (Tokyo) 51: production. Am J Respir Cell Mol Biol 36: 244‑253, 2007. 467‑470, 2003. 4. Zhu L, Lee PK, Lee WM, Zhao Y, Yu D and Chen Y: 25. Mao GX, Wang Y, Qiu Q, Deng HB, Yuan LG, Li RG, Song DQ, Rhinovirus‑induced major airway mucin production involves a Li YY, Li DD and Wang Z: Salidroside protects human fibroblast novel TLR3‑EGFR‑dependent pathway. Am J Respir Cell Mol cells from premature senescence induced by H(2)O(2) partly Biol 40: 610‑619, 2009. through modulating oxidative status. Mech Ageing Dev 131: 5. Turner J and Jones CE: Regulation of mucin expression in respi- 723‑731, 2010. ratory diseases. Biochem Soc Trans 37: 877‑881, 2009. 26. Li X, Ye X, Li X, Sun X, Liang Q, Tao L, Kang X and Chen J: 6. Caterina MJ, Leffler A, Malmberg AB, Martin WJ, Trafton J, Salidroside protects against MPP(+)‑induced apoptosis in PC12 Petersen‑Zeitz KR, Koltzenburg M, Basbaum AI and Julius D: cells by inhibiting the NO pathway. Brain Res 1382: 9‑18, 2011. Impaired nociception and pain sensation in mice lacking the 27. Zhang L, Yu H, Sun Y, Lin X, Chen B, Tan C, Cao G and Wang Z: capsaicin receptor. Science 288: 306‑313, 2000. Protective effects of salidroside on hydrogen peroxide‑induced 7. Caterina MJ, Rosen TA, Tominaga M, Brake AJ and Julius D: A apoptosis in SH‑SY5Y human neuroblastoma cells. Eur J capsaicin‑receptor homologue with a high threshold for noxious Pharmacol 564: 18‑25, 2007. heat. Nature 398: 436‑441, 1999. 28. Koskela HO: Cold air‑provoked respiratory symptoms: the 8. Smith GD, Gunthorpe MJ, Kelsell RE, Hayes PD, Reilly P, mechanisms and management. Int J Circumpolar Health 66: Facer P, Wright JE, Jerman JC, Walhin JP, Ooi L, Egerton J, 91‑100, 2007. Charles KJ, Smart D, Randall AD, Anand P and Davis JB: 29. Rodway GW and Windsor JS: Airway mucociliary function at TRPV3 is a temperature‑sensitive vanilloid receptor‑like protein. high altitude. Wilderness Environ Med 17: 271‑275, 2006. Nature 418: 186‑190, 2002. 30. Dada LA and Sznajder JI: Mitochondrial Ca2+ and ROS take 9. Güler AD, Lee H, Iida T, Shimizu I, Tominaga M and Caterina M: center stage to orchestrate TNF‑α‑mediated inflammatory Heat‑evoked activation of the ion channel, TRPV4. J Neurosci 2: responses. J Clin Invest 121: 1683‑1685, 2011. 6408‑6414, 2002. 31. Gusarova GA, Trejo HE, Dada LA, Briva A, Welch LC, 10. McKemy DD, Neuhausser WM and Julius D: Identification of a Hamanaka RB, Mutlu GM, Chandel NS, Prakriya M and cold receptor reveals a general role for TRP channels in thermo- Sznajder JI: Hypoxia leads to Na,K‑ATPase downregulation via sensation. Nature 416: 52‑58, 2002. Ca(2+) release‑activated Ca(2+) channels and AMPK activation. 11. Peier AM, Moqrich A, Hergarden AC, Reeve AJ, Andersson DA, Mol Cell Biol 31: 3546‑3556, 2011. Story GM, Earley TJ, Dragoni I, McIntyre P, Bevan S and 32. Galli GL, Lipnick MS, Shiels HA and Block BA: Temperature Patapoutian A: A TRP channel that senses cold stimuli and effects on Ca2+ cycling in scombrid cardiomyocytes: a phylo- menthol. Cell 108: 705‑715, 2002. genetic comparison. J Exp Biol 214: 1068‑1076, 2011. 646 LI et al: SALIDROSIDE INHIBITS TRPM8 ACTIVATION

33. Cho Y, Jang Y, Yang YD, Lee CH, Lee Y and Oh U: TRPM8 40. Bavencoffe A, Gkika D, Kondratskyi A, Beck B, Borowiec AS, mediates cold and menthol allergies associated with mast cell Bidaux G, Busserolles J, Eschalier A, Shuba Y, Skryma R and activation. Cell Calcium 48: 202‑208, 2010. Prevarskaya N: The transient receptor potential channel TRPM8 34. Abe J, Hosokawa H, Sawada Y, Matsumura K and Kobayashi S: is inhibited via the alpha 2A adrenoreceptor signaling pathway. J Ca2+‑dependent PKC activation mediates menthol‑induced Biol Chem 285: 9410‑9419, 2010. desensitization of transient receptor potential M8. Neurosci 41. De Petrocellis L, Starowicz K, Moriello AS, Vivese M, Orlando P Lett 397: 140‑144, 2006. and Di Marzo V: Regulation of transient receptor potential 35. Premkumar LS, Raisinghani M, Pingle SC, Long C and channels of melastatin type 8 (TRPM8): effect of cAMP, canna- Pimentel F: Downregulation of transient receptor potential binoid CB(1) receptors and endovanilloids. Exp Cell Res 313: melastatin 8 by protein kinase C‑mediated dephosphorylation. J 1911‑1920, 2007. Neurosci 25: 11322‑11329, 2005. 42. Andrisani OM: CREB‑mediated transcriptional control. Crit 36. Daniels RL, Takashima Y and McKemy DD: Activity of the Rev Eukaryot Gene Expr 9: 19‑32, 1999. neuronal cold sensor TRPM8 is regulated by phospholipase C 43. Johannessen M and Moens U: Multisite phosphorylation of the via the phospholipid phosphoinositol 4,5‑bisphosphate. J Biol cAMP response element‑binding protein (CREB) by a diversity Chem 84: 1570‑1582, 2009. of protein kinases. Front Biosci 12: 1814‑1832, 2007. 37. Rohács T, Lopes CM, Michailidis I and Logothetis DE: PI(4,5)P2 44. Johannessen M, Delghandi MP, Seternes OM, Johansen B and regulates the activation and desensitization of TRPM8 channels Moens U: Synergistic activation of CREB‑mediated transcription through the TRP domain. Nat Neurosci 8: 626‑634, 2005. by forskolin and phorbol ester requires PKC and depends on 38. Andersson DA, Nash M and Bevan S: Modulation of the the glutamine‑rich Q2 transactivation domain. Cell Signal 16: cold‑activated channel TRPM8 by lysophospholipids and poly- 1187‑1199, 2004. unsaturated fatty acids. J Neurosci 27: 3347‑3355, 2007. 45. Shaywitz AJ and Greenberg ME: CREB: a stimulus-induced 39. Vanden Abeele F, Zholos A, Bidaux G, Shuba Y, Thebault S, transcription factor activated by a diverse array of extracellular Beck B, Flourakis M, Panchin Y, Skryma R and Prevarskaya N: signals. Annu Rev Biochem 68: 821‑861, 1999. 2+ 46. Zhang L and Barritt GJ: Evidence that TRPM8 is an Ca ‑independent phospholipase A2‑dependent gating of 2+ TRPM8 by lysophospholipids. J Biol Chem 281: 40174‑40182, androgen‑dependent Ca channel required for the survival of prostate cancer cells. Cancer Res 64: 8365‑8373, 2004. 2006.